Finding the missing piece of the aquatic plastic pollution puzzle: Interaction between primary producers and microplastics

Abstract

Mounting evidence of global plastic pollution has prompted many studies of its potential effects on aquatic ecosystems. In particular, most research has focused on organismal responses to microplastics and the effects of microplastics when introduced as food. However, there has been far less research into the possible effects of microplastics on primary producers. In this review, we document the available evidence for possible effects from the literature and from a laboratory experiment using cyanobacteria and microplastic fragments. Our review shows that primary producer–microplastic interactions can alter algal photosynthesis, growth, gene expression, and colony size and morphology, possibly via adhesion and/or transfer of adsorbed pollutants from microplastics; and, these effects could be transferred up the food web, including to humans. We recommend that the effects of microplastics on primary producers be incorporated into microplastic research to better understand its full effects on freshwater and marine ecosystems.

---

Scientific Significance Statement

A significant amount of plastic waste enters rivers, lakes, and oceans, where larger pieces eventually break down into smaller pieces called microplastics. Many researchers have investigated how these microplastics can be ingested by invertebrate and vertebrate animals and can lead to negative consequences for organisms and food webs. However, there is only a limited amount of research on how microplastics could alter growth patterns of primary producers, including micro- and macro-algae and aquatic plants. In this article, we summarize the available evidence from the literature and an experimental study to show that photosynthetic organisms do interact with microplastics, with measurable changes in their growth. Future research on microplastics should pay closer attention to possible influences of microplastics at the bottom of the aquatic food web.

---

Background and emerging concepts

The use of plastics worldwide has soared in the last half century and is still increasing: global plastic production (excluding certain fibers) was a mere 1.7 million tonnes in 1950 but increased to 47 million tonnes, 204 million tonnes, and 322 million tonnes in 1976, 2002, and 2015, respectively (PlasticsEurope ). Development and implementation of highly effective reuse and recycling of plastics, however, is still in its infancy. For example, only 14% (by mass) of plastic packaging materials produced globally in 2013 was collected for recycling, while 72% entered landfills or was otherwise released into the environment (World Economic Forum ).

All plastics are originally manufactured on land, and most are used on land; however, waste plastics that do not enter appropriate reuse or recycling loops on land eventually are released into rivers and coastal waters. Direct, intentional dumping into offshore waters is also probable yet difficult to document and quantify. Some virgin plastic pellets are also accidentally spilled during maritime transport (United States Environmental Protection Agency ), and abandoned fishing gear (Andrady ) also contributes to marine plastic pollution. Plastic debris of various sizes are routinely collected not only near the coasts but also in pelagic and deep oceans (Rochman et al. ). Such findings should be treated as terminal symptoms of global plastic pollution; they indicate the severity of the issue but offer limited insight on how to address the problem before it spreads from the primary site (land) to the secondary site (freshwater and marine ecosystems).

Recent reviews indicate that microscopic plastic debris are indeed ubiquitous around the world's oceans (Moore ; Law and Thompson ). In addition to this, evidence from freshwater systems is also mounting (Wagner et al. ; Eerkes-Medrano et al. ; Van Cauwenberghe et al. ). These microscopic plastic debris often result from fragmentation of larger plastic debris that enter rivers and eventually reach the oceans (Morritt et al. ). The fraction of plastic debris that are < 5 mm in the largest dimension or diameter are commonly referred to as microplastics (e.g., Arthur et al. ; United Nations Environment Programme ).

While we certainly need to improve our abilities to accurately estimate and monitor the quantity and spatial distribution patterns of marine plastic debris, the scientific community is also increasingly aware of the research needs of systems that are proximal to the sources, such as freshwater bodies (Wagner et al. ; Eerkes-Medrano et al. ) and heavily populated coastal areas where major rivers and waste treatment plant effluents enter the oceans (Fendall and Sewell ).

Primary producers comprising the base of marine and freshwater food webs, similarly, have received relatively little attention in plastic pollution research (Ivar do Sul and Costa ; Eerkes-Medrano et al. ). Studies related to microbial biofilms on plastic debris in water, including microplastics, are being published at an increasing frequency in recent years (Muthukumar et al. ; Carson et al. ; Zettler et al. ; Harrison et al. ; McCormick et al. ; Oberbeckmann et al. ). These biofilms were taxonomically dominated by heterotrophic bacteria, yet cyanobacteria were repeatedly detected as a component. We therefore postulate that microscopic primary producers are potentially an important (or even essential) constituent of biotic communities associated with aquatic microplastics. Primary producers as a group have received relatively little attention with regards to global plastic pollution; however, a handful of publications have started to elucidate the influences of micro- and nanoplastics on microalgae through laboratory experiments (Besseling et al. ; Lagarde et al. ; Sjollema et al. ) and marine plastic debris (> 2 mm) serving as net autotrophic “hot spots” in the oligotrophic North Pacific (Bryant et al. ).

In this Current Evidence article, we summarize what is known and not known about the effects of microplastics on primary producers in freshwater and marine ecosystems (Table ). We also present evidence from a laboratory study on cyanobacteria. We show that there is ample evidence that microplastics can and do interact with aquatic primary producers, and their consequences can be ecologically relevant, especially in combination with other anthropogenic effects such as climate change, eutrophication, and food web alteration.

Synthesis table: reported effects of microplastics on aquatic primary producers.

Response variable	Effect	Unicellular microalgae	Colonial or filamentous microalgae	Macrophytes
				Macroalgae
Photosynthesis	Reduced	Chlorella (FW) 0.02 μm PS beads at 1.8–6.5 mg L−1 Bhattacharya et al. ()	Scenedesmus (FW) 0.02 μm PS beads at 1.8–6.5 mg L−1 Bhattacharya et al. ()
	No effect	Dunaliella (MAR), Chlorella (FW), Thalassiosira (MAR) 0.05 μm (uncharged), 0.5 μm (uncharged and negatively charged) and 6 μm (uncharged) PS beads 25 and 250 mg L−1 Sjollema et al. () (MAR and FW)
Growth rate (cell count based)	Reduced	Dunaliella (MAR) 0.07 μm PS at 1 g L−1 Besseling et al. ()
	Reduced	Dunaliella (MAR) 0.05 μm (uncharged) and 0.5 μm (uncharged and negatively charged) PS beads at 250 mg L−1 Greater growth rate reduction with smaller beads Sjollema et al. ()
	No effect	Chlamydomonas (FW) Milled HDPE fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. () (FW)
	Reduced	Chlamydomonas (FW) Milled PP fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. ()
Expression of chloroplast genes	rbcL significantly reduced with HDPE; no effect on psaA, psaB, psbA, psbL, and petB	Chlamydomonas (FW) Milled HDPE and PP fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. () (FW)
Expression of genes involved in stress response	No effect on sod, perox, and casp	Chlamydomonas (FW) Milled HDPE and PP fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. () (FW)
Expression of genes involved in extracellular polysaccharide biosynthesis	> 700-fold increase in UGD (→ xylose) and UGE (→ galactose), HDPE effect > PP effect Threefold to fivefold decrease in UGLD (→ rhamnose), HDPE effect > PP effect No effect on PG (→ glucose)	Chlamydomonas (FW) Milled HDPE and PP fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. ()
Microplastic-biomass attachment	Present (adsorption of nanoplastics onto microalgae)	Chlorella (FW) 0.02 μm PS beads at 0.08–0.8 mg mL−1 (4 mg mL−1 for SEM) Bhattacharya et al. ()	Scenedesmus (FW) 0.02 μm PS beads at 0.08–0.8 mg mL−1 (4 mg mL−1 for SEM) Bhattacharya et al. ()
	Present (microalgae-microplastic aggregate formed, with accelerated sinking of the incorporated microplastics)	Rhodomonas (MAR) 2 μm PS beads at 104 beads mL−1 Long et al. ()	Chaetoceros (MAR) 2 μm PS beads at 104 beads mL−1 Long et al. ()
	Present with PP (microalgae-microplastic aggregate present, overall density ∼ 1.2 g mL−1)	Chlamydomonas (FW) Milled HDPE and PP fragments (400–1000 μm) at 400 mg L−1 Lagarde et al. ()
	Present (adhesion to macroalgal surface)			Fucus (MAR) 1–100 μm milled PS fragments (45,000 fragments mL−1), 10 μm PS beads (55,000 beads mL−1), and 90–2200 μm long polyacryl fibers (initial concentration unknown) Gutow et al. ()

FW, freshwater; HDPE, high density polyethylene; MAR, marine; PP, polypropylene; PS, polystyrene; UGD, UDP-glucuronate decarboxylase; UGE, UDP-glucose 4-epimerase; UGLD, UDP-glucose 4, 6 dehydratase. Gray cells indicate no widely published data for the subject area.

³Thalassiosira pseudonana, which can form chains of connected cells. No colony formation was mentioned in this study.

⁴Authors pointed out that the negative growth may have been confounded by formation of algal-PP particle aggregates; algal cells in these aggregates were not counted.

⁵No published data were available for aquatic plants.

Types of microplastics and “microbeads”

Plastics in aquatic environments vary greatly in size: macroscopic pieces tend to be called plastic debris or waste, while particles that are smaller than 5 mm are often called microplastics (Arthur et al. ). The latter can further be classified into different shape classes such as fragment, film, fiber, foam, and pellet (Free et al. ). As expected, there are many types of plastics with various chemical compositions, and identification of the exact material is not always possible for every particle collected in an environmental sample. Polypropylene (PP) and polyethylene (PE) appear to be two dominant classes of marine microplastics (Andrady ; Zettler et al. ), reflecting their ubiquitous application as packaging materials. All scientific-grade microplastic spheres used in controlled experiments reviewed here were made of PE or polystyrene (PS). These engineered microplastic spheres as well as virgin plastic pellets (1–5 mm) are called primary microplastics; those resulting from fragmentation of larger pieces of plastics are referred to as secondary microplastics (Arthur et al. ). Some plastic packaging items such as disposable carrier bags are now formulated to be degradable under controlled test protocols, but their effectiveness as a waste management method may be limited as the degradation process can release numerous microplastic particles as well as associated chemicals (e.g., polymer additives to promote oxidative decomposition) into the environment (O'Brine and Thompson ). There also is a question of whether those test protocols are realistic representation of biodegradation environment in situ at the end of the product life; for example, household composting as opposed to industrial composting (Song et al. ).

Plastic “microbeads” or scrub particles in personal care products such as face and body wash are a type of primary microplastics released into wastewater (Gregory ; Fendall and Sewell ; Cluzard et al. ; Napper et al. ). National and regional bans on products containing them are positive steps toward reducing aquatic plastic pollution. These personal care products were first studied in the 1990s: Zitko and Hanlon () found PS fragments in products sold in Canada, and Gregory () detected PE fragments in those marketed in New Zealand. Both described the particles as variously sized and shaped with sharp edges, not spheres. Fendall and Sewell () detected both spherical and irregularly shaped PE particles in facial cleansers sold in New Zealand. We made similar observations with three major brands of face and body washes sold in the U.S. as of the summer of 2014 (Fig. shows larger microplastics from one body wash) and no longer consider “microbeads” an appropriate general term to collectively address all types of plastic scrubbing particles found in personal care products. Researchers as well as the public should be aware that microplastics in water bodies include a range of shapes and sizes. When experiments only include uniformly sized and shaped analytic-grade microbeads, there is a misrepresentation of the full range of microplastics found in water bodies (Eriksen et al. ,b; Free et al. ; Mazurais et al. ).

Photomicrographs of variously shaped and size microplastics harvested from a body wash product (Product A).

Biofilm formation on microplastics: Plastisphere

Biological fouling likely plays a major role in the processes leading to long-term sequestration of plastic debris in aquatic environments (Barnes et al. ), while fouling by highly buoyant microorganisms (e.g., cyanobacteria with gas vacuoles) may temporarily increase the overall buoyancy of the plastic item until they die off or are outcompeted or consumed by organisms with higher density. While major classes of plastics encountered as marine plastic debris vary in their specific gravity (Andrady ), Thompson et al. () found a similar assortment of plastics between sediment samples and archived plankton samples from the North Atlantic. This suggests that even floating plastic debris eventually loses buoyancy due to biofouling and/or fragmentation and sink into the sediment. Plastic debris found in the surface water is literally a tip of the iceberg—a temporary and partial manifestation of the total quantity of plastic wastes accumulating in the ocean.

The term plastisphere refers to the community of biofouling organisms on plastics (Zettler et al. ). One of the earliest studies on this topic was a series of laboratory experiments by Zobell (), which found that organic matter concentrated on solid surfaces in seawater and provided a new habitat for sessile bacteria, forming a biofilm. Two further marine plastisphere research articles were published in 1972. Carpenter and Smith () characterized microplastics collected in a neuston net deployed in Sargasso Sea and documented a variety of hydroids and diatoms colonizing the plastic particles. Carpenter et al. () found microscopic plastic spheres in plankton tow samples in the coastal water of New England, USA, which were identified as PS with gram-negative rod-shaped bacteria and polychlorinated biphenyls. Deng et al. () discovered that a filamentous freshwater cyanobacterium Phormidium foveolarum (= Leptolyngbya foveolara) degraded hydrocarbons. Two more recent studies detected molecular signatures of the genus Phormidium and related hydrocarbon-degrading taxa associated with plastics: Zettler et al. () on microplastics collected from the North Atlantic and Oberbeckmann et al. () on polyethylene terephthalate (PET) bottles incubated in the North Sea. This raises an interesting possibility that Phormidium in plastisphere may be actively hydrolyzing the plastic rather than depending upon inorganic nutrients released by heterotrophic bacteria (i.e., earlier colonizers of the microplastic particle) or present in the wider environment. Some heterotrophic bacteria in plastispheres are of known hydrocarbon-degrading taxa; Harrison et al. () found them on low-density PE (LDPE) microplastics incubated in coastal marine sediment and McCormick et al. () on waste microplastics collected from a highly urbanized river. In both studies, microplastic particles harbored less diverse bacterial flora dominated by the hydrocarbon-degraders than natural substrates (e.g., sediment and suspended organic matter). Zettler et al. () showed images of numerous pits on marine microplastic surfaces that match the outline of attached bacteria, and Bryant et al. () reported (prokaryotic or eukaryotic) cells that were nested within pores of larger (> 5 mm) marine plastic particles. These observations suggest active hydrolysis of the polymer by the bacteria.

Formation of the plastisphere can significantly alter physical properties of the microplastics (e.g., size, density, and sinking rate) as well as biological properties such as detectability and likelihood of ingestion and rejection by filter feeders. We therefore anticipate that future studies of microplastic ingestion by aquatic animals will start to incorporate nonspherical secondary microplastics with plastisphere in addition to brand-new analytical grade microbeads commonly used in such studies to date.

Primary producers as plastisphere members

While heterotrophic bacteria tend to be the focus of plastisphere research, the presence of both prokaryotic and eukaryotic autotrophs within the biofilm have been documented (Muthukumar et al. ; Zettler et al. ; Oberbeckmann et al. ; Bryant et al. ). Most commonly detected are diatoms, and those in Zettler et al. () were known biofilm-forming taxa in conventional aquatic environments. This suggests that these diatoms attach to and colonize plastics through similar mechanisms as with rocks, sediments, plants, and detritus. Pre-existing bacterial biofilm was not a prerequisite for marine diatoms to attach to glass plates (Kawamura et al. ). While the biological process of diatom adhesion to microplastics is yet to be studied in detail, the general process of biofilm formation that begins with electrostatic attraction and repulsion between surfaces and negatively charged bacterial cell wall (as reviewed by Renner and Weibel ) may apply to diatoms; diatom cell surfaces have been reported as negatively charged (Edzwald ; Konno ; Gélabert et al. ). Aside from diatoms, a wide variety of other eukaryotic microalgae have been detected via DNA sequencing of plastisphere samples, including prasinophytes, rhodophytes, cryptophytes, haptophytes, dinoflagellates, (including potentially harmful Alexandrium), chlorarachniophytes, chrysophytes, pelagophytes, and phaeophytes (Zettler et al. ; Reisser et al. ).

Incubation experiments by Oberbeckmann et al. () showed that PET bottle surfaces promoted different types of microbial communities than those on glass and in ambient seawater; for example, colony-forming Stanieria and filamentous Pseudophormidium, both cyanobacteria, were detected on PET while they were below detection on glass or in seawater; the latter were dominated by Proteobacteria instead. Oberbeckmann et al. () also collected particles (0.5–10 mm diameter) in the North Sea via manta trawls and found cyanobacteria as a major constituent of the LDPE and PP plastispheres; biofilms on particles composed of chitin, paint, or a PP/high-density PE/ethylene vinyl alcohol blend were characterized by Bacteroidetes and Proteobacteria, and no signs of cyanobacteria were detected. Zettler et al. () found filamentous cyanobacteria on marine microplastics (PE and PP) but not in ambient seawater, which was instead dominated by a unicellular cyanobacterium, Prochlorococcus. Cyanobacteria were among the most common prokaryotes found on marine plastic particles (> 2 mm) from the North Pacific Gyre, with the most abundant taxa closely related to the filamentous genera including Phormidium (Bryant et al. ). The authors also detected Alphaproteobacteria on glass and in ambient water but not on plastic; it is possible that filamentous cyanobacteria somehow replaced Alphaproteobacteria on plastic. Taken together, colony-forming and filamentous cyanobacteria appear to be an important constituent of aquatic biofilms on PET, PE, and PP but not necessarily on other surfaces. More microbial community composition data comparing biofilms on different materials are needed to determine whether (1) plastics selectively promote non-unicellular cyanobacteria within the biofilm or (2) successful formation of aquatic biofilms on solid surfaces in general is aided by colony-forming and filamentous cyanobacteria.

Interaction between cyanobacteria and microplastics from a personal care product

To start addressing the abovementioned knowledge gap on how microplastics interact with aquatic primary producers and change their population dynamics, we designed a laboratory experiment to investigate impacts of microplastics on cyanobacterial abundance, biovolume, and colony size over a 21-d period. We had previously harvested and characterized microplastics and plant-based scrubbing particles from five non-antibacterial body and face wash products that were widely available in the U.S. in 2014 and decided to use microplastics from a body wash product that contained only one type of microplastics (i.e., all irregularly shaped sized particles with uniform color and density; not mixed with flexible and breakable spheres or “bursting beads”) in this experiment. Five milligrams of the non-spherical microplastics were added to 75 mL monocultures of two harmful algal bloom (HAB) forming freshwater cyanobacteria, Microcystis aeruginosa and Dolichospermum flos-aquae. Detailed methods are described in Supporting Information.

We applied the microplastics at 66.7 mg L⁻¹, which was between the two concentrations (25 mg L⁻¹ and 250 mg L⁻¹ of 0.05 μm, 0.5 μm, and 6 μm beads) in Sjollema et al. () and was 1/6 of the 400 mg L⁻¹ of milled plastic particles (400–1000 μm) used by Lagarde et al. (). Ecological relevance of these mass-based concentrations applied in laboratory studies are difficult to evaluate at this time as most field data are limited to larger (> 333 μm) microplastics and reported as numbers of microplastic particles per area or volume (e.g., Eerkes-Medrano et al. ). The majority of the microplastics used in the two studies above as well as ours (97% of our particles were < 200 μm with few larger ones) would have been missed by the standard 333 μm sampling mesh used in many field studies. For reference, the global oceanic floating plastic pollution model by Eriksen et al. () predicted relatively high concentration of 333–1000 μm microplastics around the Hawaiian Archipelago, U.S.A., at ca. 1000 g km⁻². A crude assumption of the mean depth of 5000 m for the area and vertically uniform distribution of microplastics yields 0.2 mg L⁻¹ as the mean concentration for this larger size class. The modeling results, however, suggested differential loss of small particles from the surface water (Eriksen et al. ), which may concentrate smaller microplastics into the deepest part of the water column over time up to levels comparable to those in the studies above.

Microcystis forms mucilaginous colonies when blooming in lakes and ponds; however, colony formation was not detected in our experimental or stock cultures, consistent with its morphological plasticity described in literature (Fulton and Paerl ; Doers and Parker ). Within the 21-d period, we observed multiple signs of interaction between these two cyanobacteria and added microplastics: in Microcystis, algal particle counts (per mL) were higher with microplastics on Days 3 and 12 (Fig. a; t = −2.81, df = 6.80, p = 0.027 for Day 3; t = −2.10, df = 7.58, p = 0.071 for Day 12). In Dolichospermum cultures, algal particle counts were higher with microplastics on Day 9 (t = −2.16, df = 6.06, p = 0.074). In both cyanobacteria, significantly smaller algal particle size was recorded for the microplastic-treated cultures at one point during the experiment (Fig. e,f, Day 3 for Microcystis (t = 2.48, df = 6.91, p = 0.043) and Day 9 for Dolichospermum (t = 4.09, df = 4.03, p = 0.015). These signs of algal-microplastic interaction did not persist throughout the experimental period, implying a dynamic interaction between microplastics and algal growth rates and life cycles. This was partially supported by the statistically significant microplastic × day effect in Dolichospermum cultures (p = 0.023, F_4,20 = 3.608, repeated measures analysis of variance [ANOVA]). Their Day 9 data indicated highly synchronous filament elongation in the control cultures (Fig. f). Microplastic-treated Dolichospermum cultures, in contrast, lacked such filament elongation. In both cyanobacteria, microplastic treatment had no sustained effect on algal biomass and growth (as measured as total algal particulate volume) throughout the 21-d period (Fig. c,d), and the effects of microplastics were detected as smaller algal particle sizes. These results practically ruled out the possibility that the observed effects were due to simple growth inhibition by the body wash (e.g., surfactant) residues remaining on the microplastics, as such effects should exhibit a more consistent pattern over the experimental duration. Visual inspection of the microplastics further revealed that the filamentous Dolichospermum often formed bundles attached to corners of irregularly shaped microplastics, while unicellular Microcystis did not show such patterns of adhesion (Fig. ). The algal-plastic adhesion discussed here was observed between algal cells that were < 10 μm in the largest dimension as unicells and microplastics that were typically 10–100 times larger than the algae, possibly providing a favorable surface for colony establishment and/or maintenance. This contrasts with studies using nanoplastic beads that were up to 100–1000 times smaller than algal cells, where nanoplastics may cling to the algal cell and potentially reduce cell surface area available for processes such as light capture and diffusion. Bhattacharya et al. () reported reduced photosynthesis in Chlorella and Scenedesmus (both in Chlorophyta) treated with 0.02 μm PS beads, while Sjollema et al. () studied two chlorophytes (Dunaliella and Chlorella) and one diatom (Thalassiosira) treated with 0.05 μm, 0.5 μm, and 6 μm PS beads and observed reduced algal growth at high concentration of smaller nanoplastics, yet no direct effect on photosynthetic rate.

Interaction between microplastics from a body wash and two cyanobacteria, Microcystis arruginosa (a, c, e) and D. flos-aquae (b, d, f) in a 21-d batch culture experiment. ESD, equivalent spherical diameter. Open circles: control; dark squares: microplastic treatment. N = 5 for each species-treatment combination. Error bars represent ± 1 SE. Note: the particle analysis did not distinguish between algal and microplastic particles; however, the direct contribution of the added microplastics (∼ 1200 particles mL−1, ∼ 1.1 × 106 μm3 mL−1, and ESD ≫ 20 μm) on the three responses (# of particles, total particulate volume, and median ESD) was negligible in comparison with that of the algae themselves.

Representative images of microplastics in (a) Microcystis and (b) and (c) Dolichospermum cultures after 21-d.

Ecological implications of algal-microplastic interaction

Our results above, most likely the first for freshwater cyanobacteria, are consistent with the earlier observations in marine studies that identified colonial and filamentous cyanobacteria as an important component of the plastisphere (Zettler et al. ; Oberbeckmann et al. ). As colonial and filamentous cyanobacteria are frequently the cause of HAB in inland waters (Lee ), their interaction with microplastics remaining in wastewater effluent can potentially affect HAB formation and persistence. For example, a filamentous cyanobacterium flourishing within the plastisphere of floating microplastics can aggregate and increase surface scum formation. Attached cyanobacteria may also slow down sequestration of microplastics into the sediment; many HAB-forming cyanobacteria increase their buoyancy via gas vacuoles (Oliver ). This may confer decreased density to the entire particle (a microplastic plus plastisphere). On the other hand, filamentous cyanobacteria attached and entangled onto variously shaped microplastics, often with sharp edges (Gregory ), may lead to fragmentation of longer algal filaments and result in more dispersed growth patterns than clumps and mats. Specialized cyanobacterial cells such as heterocysts (for nitrogen fixation) and akinetes (for semi-dormancy during cold periods) do not have gas vacuoles, and when attached to neutrally buoyant or sinking microplastics, may increase sinking of microplastics. Cyanobacterial cells hitchhiking on rapidly sinking microplastics can also provide a novel path for them to escape unfavorable growth conditions that may exist in surface water, such as low pH, self-shading, and nutrient limitation (Lee ). While the vegetative cells will die when permanently in an aphotic zone, some that get detached from the sinking microplastics may be resuspended into the photic zone and flourish when growth conditions become favorable again. Cyanobacterial akinetes can overwinter in the sediment-water interface and later germinate into vegetative cells (Lee ). Akinete hitchhiking on sinking microplastics may therefore increase the size of cyanobacterial “seed bank” in the sediment and possibly the frequency and/or severity of future HABs.

Few other studies have been published on interaction between microscopic plastics and algae. Bhattacharya et al. () studied effects of plastic nanoparticles (ca. 20 nm diameter) on photosynthesis by two chlorophytes, Chlorella and Scenedesmus. The nanoplastics were positively or negatively charged at the surface, and the algae were 2–3 orders of magnitude larger (2–78 μm in the largest dimension) than the nanoparticles. Positively charged nanoparticles showed more adsorption to the algal cells than negatively charged nanoparticles did, which curiously was consistent with Zobell () that found positively charged surfaces attracted more bacteria. The commonality between the two observations of positively charged plastics, regardless of size, promoting adhesion of microorganisms lead us to hypothesize that surface charge of primary producers is a key factor for microplastic attachment. Further investigation of this topic will also enable us to systematically evaluate if a scaling effect exists between the size of plastics and algal attachment while controlling for the charge differential. The nanoplastics that adsorbed to algal cells in Bhattacharya et al. () lead to decreased photosynthesis and increased production of reactive oxygen species, indicating reduced photosynthetic efficiency. Furthermore, in the same study, colonial Scenedesmus attracted more nanoplastics than unicellular Chlorella did. This is a chlorophyte parallel to the three independent accounts of filamentous or colonial cyanobacteria exhibiting higher affinity toward microplastics than unicells—Zettler et al. (); Oberbeckmann et al. (); and our laboratory experiment above.

Macroalgae and aquatic plants

While much of our discussion above on the interaction between microplastics and primary producers focused on microalgae, microplastics may affect biology of macroscopic algae (macroalgae) either directly or indirectly. Gutow et al. () reported that suspended PS microplastics collected on the surfaces of a brown alga Fucus vesiculosus were consumed by the common periwinkle, Littorina littorea, along with the algal biomass. They reported a high variability in microplastic particle concentration found on algal surfaces and suggested algal morphology, mucus, and electrostatic interaction as major factors for microplastic adhesion. While the authors did not find direct negative effects of microplastics on the algae nor the periwinkle in the study, they pointed out the possibility of the microplastics serving as a vector of chemical contaminants into the macroalgal biomass and its consumers.

Macroalgae typically have microscopic life stages as gametes, spores, and developing thalli (Lee ), which are often morphologically similar to microalgae. This also applies to aquatic vascular plants that reproduce sexually via gametes released into water or asexually via planktonic resting stages or microscopic plant fragments. Chan et al. () showed that brown algae (Sargassum spp.) efficiently accumulated di(2-ethylhexyl) phthalate, a persistent organic pollutant (POP) used as a PVC additive, in their biomass. While the majority of published data on reproductive effects of POPs are from humans and other mammals, there are recent reports of reduced fertilization and increased developmental defects in an invertebrate (Pacific oyster, Crassostrea gigas) (Mai et al. ) and developmental inhibition in rape plant (Brassica rapa L., formerly known as B. chinensis L.) seedlings (Ma et al. ), both observed at environmentally relevant concentrations. These studies suggest that POPs can disrupt successful reproduction of simpler organisms that lack sophisticated endocrine systems. We thus conjecture that POPs adsorbed onto microplastics (Andrady ; Napper et al. ) have potential to rapidly translocate into aquatic macrophyte biomass and, when applicable, disturb alternation of sexual and asexual reproductive cycles in addition to general growth inhibition. As macrophytes are sessile for the majority of their life span and are less likely to successfully escape from local diseases and climatological anomalies, mismatch between the sexual-asexual control switch and the environmental cues can have greater implications on their genetic diversity and resilience to disturbances than for planktonic microalgae. These macrophytes provide important ecological services in coastal areas and littoral zones of freshwater systems, such as sediment stabilization and providing habitats for epifauna and infauna. Some of these habitats serve as prime feeding grounds and nurseries for juvenile vertebrates, and their increased exposure to microplastic-derived POPs in or on macrophytes via direct contact and ingestion may have significant implications on their development, survival, and future reproduction. Holistic evaluation of the effects of microplastics in these ecosystems, therefore, warrants consideration of direct microplastic-macrophyte interactions.

Human health implications

Current discussion of how aquatic microplastic pollution can negatively affect human health is centered around (1) bioaccumulation and transfer of microplastics (Van Cauwenberghe and Janssen ; Avio et al. ) and associated POPs (Avio et al. ) via shellfish consumption by humans and (2) detection of potentially pathogenic bacteria (e.g., genus Vibrio) on microplastic surfaces (Zettler et al. ; Kirstein et al. ). Vibrio often colonizes biotic surfaces such as those covered by phytoplankton, zooplankton, or corals (Kirstein et al. ), which suggests that biofilm-coated microplastics are more likely to harbor Vibrio than bare microplastics. This can potentially increase the risk of Vibrio infection in humans in or near waters heavily polluted with microplastics, through direct contact with broken skin or consumption of mollusks that are contaminated with Vibrio.

Humans, however, are omnivores that can also ingest microplastics and associated contaminants attached or incorporated into non-animal based food products. Yang et al. () detected much higher levels of microplastic contamination in table salts made of sea salts than those made of lake and rock/well salts. This empirically suggests that microplastics existed in coastal seawater and were further concentrated into the final product by evaporation. Seaweed farms around the world that harvest algae for human consumption also operate in shallow coastal waters, and macroalgae with delicate thalli, such as Porphyra, also go through concentration and dehydration after harvest before they are processed into uniformly sized sheets marketed as Nori or “sea vegetable” (Lee ). Detection and removal of microplastics in the raw material will be very difficult in this conventional production system, while large-scale culture in a closed system will be cost prohibitive for most communities engaged in seaweed farming. These farming communities are typically located close to major human settlements, which Free et al. () found to be a major driving factor for microplastic pollution even when the general geographical area is considered remote and rural. Widely used food additives such as agar and carageenan are also made from marine macroalgae harvested in the ocean (Lee ) and may be contaminated with microplastics; the cloth filters used in standard manufacturing protocols (McHugh ) are not expected to effectively remove microplastics in the micrometer range. Dietary supplements made of microalgae (e.g., Spirulina) are typically produced in a closed culture system where microplastic contamination is of lesser concern.

Future research needs

We have shown that aquatic primary producers interact with microplastics, either directly or mediated by heterotrophic bacteria, with a wide range of possible ecological consequences. We have only just begun to systematically investigate such dynamic interactions in mainstream marine and freshwater microplastic research. Here, we suggest two often overlooked aspects that should be incorporated into future studies: (1) use of microplastics that better simulate secondary microplastics found in situ and (2) possible morphological change in algae that are fed to animals in combination with microplastics.

• Better simulation of secondary microplastic (including nanoplastic) exposure

Most laboratory and mesocosm experiments have utilized highly spherical analytic-grade microplastics, often with fluorescent or other types of dyes for ease of detection. They are primary virgin microplastics that are readily available to the scientific community as calibration beads for flow cytometry and other techniques with known chemical composition and particle size, which makes them a popular choice for controlled experiments. They, however, do not mimic secondary microplastics commonly found in aquatic systems that vary in size, shape and chemical composition, and, most importantly, have some degree of biofouling and plastisphere formation. This issue bears resemblance to the recent discussion on whether laboratory studies should focus on environmentally relevant concentrations of microplastics or estimation of the entire dose-response curve that extends into microplastic concentrations that are orders of magnitude higher than typically reported from field studies (Huvet et al. ; Lenz et al. ). Highly engineered and quality controlled micro- and nanobeads are useful tools for establishing the basic patterns of organismal responses to environmental microplastics. For example, we need more conclusive research on the impacts of nanoplastics on algal photosynthesis. One study has shown decreased algal photosynthesis with nanobead exposure (Bhattacharya et al. ) whereas another reported decreased algal growth over time with a high concentration of smaller nanobeads yet no effect on algal photosynthetic rate (Sjollema et al. ). These somehow conflicting observations may potentially be driven by physical and physiological dynamics between algae and microplastics that scale with their sizes. To test such a hypothesis, it is only logical to keep using uniformly sized and shaped primary nanobeads until presence or absence of such scaling effect is determined. We, however, need to gradually shift our research efforts to secondary microplastics that better mimic in situ exposure of aquatic organisms to aquatic microplastics, as Sussarellu et al. () explicitly stated in their study of oysters fed with analytic-grade microbeads. We understand it is not easy; for our algal culture experiment we tried to procure variously shaped low quality primary microplastics from manufacturers who claimed to provide them to the cosmetic industry, but none of the companies that we contacted ever responded to our inquiries. There are opportunities for plastic manufacturers and the research community to develop what may eventually become the “standard secondary microplastics,” with or without freshwater or marine plastisphere, that will serve similar functions as standard strains of phytoplankters and zooplankters available from culture collections.

• 2. Consideration of possible morphological change in algae upon microplastic exposure

Mixtures of conventional food or prey and analytic-grade microbeads are often used in feeding studies to simulate exposure of aquatic animals to environmental microplastics (Carlos de Sá et al. ; Cole et al. ; Sussarellu et al. ). When microalgae are used as food, however, there typically is limited monitoring of possible morphological change in the algae over the course of the experiment. This is not trivial, as commonly used marine diatoms such as Thalassiosira and Chaetoceros can form chains. Both are species-rich genera and contain cryptic species that are difficult, if not impossible, to identify morphologically (Hamsher et al. ). While certain species such as Thalassiosira weissflogii are considered unicellular (Smayda and Boleynl ), diatoms in general can exhibit marked morphological changes in long-term cultures (Estes and Dute ). When algal cultures are mixed with microplastics, potentially with a variety of metabolites from the plastisphere and/or adsorbed POPs, any one of these compounds may act as a chemical cue to trigger morphological changes in the algae (Hessen and Van Donk ). Diatoms excrete large volumes of highly adhesive transparent exopolymer particles (Passow ), which can lead to colony or aggregate formation. Chaetoceros additionally has four long spines that extend from corners of their frustules that can cause multiple cells to link together to form a chain or an aggregate. Long et al. () produced algal aggregates by growing Chaetoceros and Rhodomonas (Cryptophyta) in rolling tanks (as monocultures and also as a mixed culture) and found that added microplastics (2 μm PS beads) were incorporated and concentrated into these aggregates, leading to accelerated sinking of the microplastics. Formation of dense (∼ 1.2 g mL⁻¹) aggregates comprised of microalgae, microplastics, and extracellular polysaccharide was also observed in Lagarde et al. () when Chlamydomonas (Chlorophyta) was grown with milled (400–1000 μm) PP, along with increased expression of certain genes involved in extracellular polysaccharide biosynthesis. These algae-microplastic aggregates are expected to be lower quality food than algae-only aggregates. Even when growing only as solitary unicells, algae mixed with microplastics may still exhibit morphological changes that could further affect their detectability, palatability, and ease of handling by animals. All of these possible outcomes of algae-microplastic interactions have implications for food web energetics that many microplastic feeding experiments try to address.

Conclusions

Scientific knowledge on the quantity, special distribution, and processes of degradation and sequestration of plastic waste has steadily advanced over the last few decades. Additionally, discovery and characterization of microbial communities that form on microplastics have opened up a new avenue to decipher the final fate of highly fragmented plastic wastes in aquatic environments. Microscopic primary producers are often a major component of the plastisphere, and macroscopic primary producers can provide a bypass in the food web through which microplastic ingestion by humans is accelerated. Interaction between primary producers and microplastics should therefore be treated as one of the high priority areas in microplastics research, and the resulting knowledge will fill one of the key knowledge gaps. Effective collaboration among multiple research fields (e.g., microbiology, organismal biology, chemistry, physics, and material science) will be needed to establish a more complete understanding of the environmental impacts and fate of plastic wastes of all sizes, which in turn is essential for successful and sustainable management of the global plastic pollution.

Acknowledgments

The co-authors thank the editors and the anonymous reviewers for their constructive comments that helped improve this manuscript and Matt Albright for technical assistance. Parts of this study were supported by SUNY Oneonta Faculty Research Grant, Biological Field Station (BFS), and BFS Summer Faculty Research Stipend. ED and BW were sponsored for the BFS Summer Internship by the Otsego Land Trust and CH by the Otsego County Conservation Association.